#importing the libraries
import os
import glob
import path

#all the libraries for models
import pandas as pd 
import numpy as np 
from sklearn.preprocessing import StandardScaler 
from sklearn.linear_model  import Ridge,Lasso,RidgeCV, LassoCV, ElasticNet, ElasticNetCV, LogisticRegression
from sklearn.model_selection import train_test_split
from statsmodels.stats.outliers_influence import variance_inflation_factor 
from sklearn.metrics import accuracy_score, confusion_matrix, roc_curve, roc_auc_score
import matplotlib.pyplot as plt
from pandas_profiling import ProfileReport
import seaborn as sns
import pickle

#logger
import logging as lg
lg.basicConfig(filename="load_data.log", level=lg.DEBUG, format="%(asctime)s, %(lineno)s, %(message)s")

#the dataset folder
try:
    lg.info("dataset folder")
    path = os.getcwd()
    print(path)
except Exception as e:
    lg.error(e)

C:\Users\sid\Downloads\AReM

ls

 Volume in drive C is Windows
 Volume Serial Number is 82A0-46B9

 Directory of C:\Users\sid\Downloads\AReM

16-09-2021  17:06    <DIR>          .
16-09-2021  17:06    <DIR>          ..
14-09-2021  22:13    <DIR>          .ipynb_checkpoints
13-09-2021  10:25    <DIR>          bending1
16-09-2021  12:44    <DIR>          bending2
18-05-2016  22:19            21,207 bendingType.pdf
18-05-2016  22:44    <DIR>          cycling
15-09-2021  12:27         4,005,910 final.ipynb
16-09-2021  17:06            31,408 final1.ipynb
16-09-2021  17:06         1,258,611 load_data.log
18-05-2016  22:50    <DIR>          lying
15-09-2021  16:45           115,054 opening.ipynb
18-05-2016  23:05           105,307 sensorsPlacement.pdf
18-05-2016  22:52    <DIR>          sitting
18-05-2016  22:53    <DIR>          standing
18-05-2016  22:55    <DIR>          walking
               6 File(s)      5,537,497 bytes
              10 Dir(s)  10,951,143,424 bytes free

try:
    lg.info("combining bending1 dataset into one ")
    path = r'C:/Users/sid/Downloads/AReM/bending1'
    all_csv = glob.glob(path + "/*.csv")
    lis = []
    for csv in all_csv:
        df = pd.read_csv(csv, header=None, sep=',', skiprows=5, error_bad_lines=False)
        lis.append(df)
    bending1 = pd.concat(lis, axis=0, ignore_index=True)
except Exception as e:
    lg.error(e)

lg.info("dataset shape")
bending1.shape

(3360, 7)

lg.info("info of dataset")
bending1.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 3360 entries, 0 to 3359
Data columns (total 7 columns):
 #   Column  Non-Null Count  Dtype  
---  ------  --------------  -----  
 0   0       3360 non-null   int64  
 1   1       3360 non-null   float64
 2   2       3360 non-null   float64
 3   3       3360 non-null   float64
 4   4       3360 non-null   float64
 5   5       3360 non-null   float64
 6   6       3360 non-null   float64
dtypes: float64(6), int64(1)
memory usage: 183.9 KB

lg.info("dataset description")
bending1.describe()

	0	1	2	3	4	5	6
count	3360.000000	3360.000000	3360.000000	3360.000000	3360.000000	3360.000000	3360.00000
mean	59875.000000	42.667723	0.454360	19.158830	0.793399	30.201982	0.55239
std	34646.097028	2.361918	0.400185	4.553348	0.915341	5.761234	0.54323
min	0.000000	33.000000	0.000000	1.500000	0.000000	8.000000	0.00000
25%	29937.500000	41.750000	0.000000	16.000000	0.000000	24.250000	0.00000
50%	59875.000000	42.750000	0.470000	19.500000	0.500000	31.500000	0.47000
75%	89812.500000	44.500000	0.500000	23.000000	1.120000	35.500000	0.87000
max	119750.000000	48.000000	3.000000	30.000000	10.030000	38.670000	5.26000

bending1.dtypes

0      int64
1    float64
2    float64
3    float64
4    float64
5    float64
6    float64
dtype: object

try:
    lg.info("adding the label in dataset")
    bending1['label'] = 'bending1'
except Exception as e:
    lg.error(e)

bending1.dtypes

0          int64
1        float64
2        float64
3        float64
4        float64
5        float64
6        float64
label     object
dtype: object

lg.info("changing the dataset folder")
try:
    os.chdir(r"C:\Users\sid\Downloads\AReM\bending2")
except Exception as e:
    lg.error(e)

try:
    lg.info("all the dataset files in bending2 folder")
    df1 = pd.read_csv('dataset1.csv', sep=',', header=None, skiprows=5)
    df2 = pd.read_csv('dataset2.csv', sep=',', header=None, skiprows=5)
    df3 = pd.read_csv('dataset3.csv', sep=',', header=None, skiprows=5)
    df4 = pd.read_csv('dataset4.csv', sep=' ', header=None, skiprows=5 ,
                 usecols =[0,1,2,3,4,5,6])
    df5 = pd.read_csv('dataset5.csv', header=None, skiprows=5)
    df6 = pd.read_csv('dataset6.csv', header=None, skiprows=5)
except Exception as e:
    lg.error(e)

data_frames = [df1, df2, df3, df4, df5, df6]
data_frames

[          0      1     2      3     4      5     6
 0         0  23.75  0.43  24.00  0.00  24.67  0.47
 1       250  22.67  0.94  24.00  0.00  24.67  0.94
 2       500  21.33  0.47  24.00  0.00  24.75  0.83
 3       750  21.00  0.00  24.00  0.00  25.25  1.09
 4      1000  23.00  0.71  23.75  0.43  24.75  0.83
 ..      ...    ...   ...    ...   ...    ...   ...
 475  118750  27.50  0.50  22.75  1.30  21.00  0.00
 476  119000  27.00  0.00  21.00  0.00  20.75  1.09
 477  119250  27.00  0.00  21.00  0.00  20.50  0.50
 478  119500  27.25  0.43  23.50  0.87  20.75  0.43
 479  119750  27.00  0.00  24.00  0.00  19.25  0.83
 
 [480 rows x 7 columns],
           0      1     2      3     4      5     6
 0         0  27.00  0.00  24.00  0.00  19.00  1.00
 1       250  27.00  0.00  24.00  0.00  19.75  0.83
 2       500  27.00  0.00  23.50  0.50  20.00  1.22
 3       750  27.00  0.00  24.00  0.00  14.75  1.48
 4      1000  27.00  0.00  23.75  0.43  14.50  1.50
 ..      ...    ...   ...    ...   ...    ...   ...
 475  118750  23.00  0.71  11.25  2.49  26.75  1.92
 476  119000  22.25  0.43  15.50  0.87  29.50  1.50
 477  119250  23.75  0.43  17.67  0.47  25.67  3.09
 478  119500  24.25  0.43  13.33  1.25  30.00  2.45
 479  119750  25.33  0.47  14.75  1.09  25.33  3.68
 
 [480 rows x 7 columns],
           0      1     2      3     4      5      6
 0         0  24.50  0.50  15.50  0.87  26.00   2.74
 1       250  25.00  0.00  13.50  3.64  33.50   2.60
 2       500  26.00  0.71  14.25  3.90  18.50  13.61
 3       750  24.33  0.47  15.67  2.62  25.50   3.64
 4      1000  23.50  0.87  16.25  1.30  23.00   1.22
 ..      ...    ...   ...    ...   ...    ...    ...
 475  118750  33.00  0.00   1.25  0.43  19.00   1.00
 476  119000  33.00  0.00   3.75  0.83  19.00   1.00
 477  119250  33.00  0.00   5.25  1.30  18.50   0.50
 478  119500  32.75  0.43   2.50  0.50  19.25   0.83
 479  119750  32.67  0.47   2.00  1.00  19.00   1.00
 
 [480 rows x 7 columns],
           0      1     2     3     4      5     6
 0         0  32.50  0.50  0.00  0.00  19.00  1.00
 1       250  32.50  0.50  0.00  0.00  18.50  0.50
 2       500  32.75  0.43  1.00  0.00  18.00  0.00
 3       750  32.50  0.50  0.00  0.00  17.50  0.50
 4      1000  32.50  0.50  7.50  0.50  17.50  0.87
 ..      ...    ...   ...   ...   ...    ...   ...
 475  118750  28.67  0.47  4.67  1.25  17.33  0.47
 476  119000  27.50  0.50  5.50  2.50  17.25  1.30
 477  119250  28.00  0.00  6.67  0.94  17.00  1.00
 478  119500  28.00  0.00  5.00  0.82  17.00  0.71
 479  119750  28.00  0.00  0.00  0.00  17.00  1.00
 
 [480 rows x 7 columns],
           0      1     2      3     4      5     6
 0         0  28.00  0.00   0.00  0.00  17.25  0.83
 1       250  29.00  0.00   0.00  0.00  17.33  0.94
 2       500  28.67  0.47   0.00  0.00  17.00  0.82
 3       750  29.00  0.00   3.00  2.00  17.67  0.47
 4      1000  29.00  0.00   3.33  0.47  18.00  1.00
 ..      ...    ...   ...    ...   ...    ...   ...
 475  118750  36.00  0.00  18.00  0.00  27.75  0.43
 476  119000  36.25  0.43  14.50  2.87  27.00  0.71
 477  119250  36.00  0.00   8.67  2.49  27.00  0.00
 478  119500  36.80  0.98  13.00  0.63  27.75  0.43
 479  119750  35.00  1.41   7.00  2.00  27.50  0.50
 
 [480 rows x 7 columns],
           0      1     2      3     4      5     6
 0         0  34.50  1.50   8.00  1.58  28.00  0.71
 1       250  36.00  0.00  11.25  3.49  28.50  0.50
 2       500  36.00  0.00  20.00  1.22  27.00  0.71
 3       750  38.25  0.43  21.25  0.43  26.00  0.71
 4      1000  37.50  1.12  22.00  0.82  23.67  2.49
 ..      ...    ...   ...    ...   ...    ...   ...
 475  118750  27.00  0.00  14.00  0.00   6.50  0.50
 476  119000  27.00  0.00  15.00  0.00   4.00  1.00
 477  119250  27.00  0.00  14.25  0.83   4.00  2.00
 478  119500  27.00  0.00  13.33  1.25   1.50  0.50
 479  119750  26.50  0.50  15.00  0.00   3.00  0.00
 
 [480 rows x 7 columns]]

try:
    lg.info("combining the dataset of bending2")
    from functools import reduce
    df_bending2 = reduce(lambda left, right:pd.merge(left, right, how='outer'),
                    data_frames)
except Exception as e:
    lg.error(e)

try:
    lg.info("adding the label in bending2 dataset files")
    df_bending2['label'] = 'bending2'
except Exception as e:
    lg.error(e)

df_bending2.dtypes

0          int64
1        float64
2        float64
3        float64
4        float64
5        float64
6        float64
label     object
dtype: object

try:
    lg.info("the dataset where cycling datasetfiles are")
    path = r'C:\Users\sid\Downloads\AReM\cycling'
    all_csv = glob.glob(path + "/*.csv")
    lis = []
    
    for csv in all_csv:
        df = pd.read_csv(csv , sep=',' , header=None, skiprows=5,  error_bad_lines=False)
        lis.append(df)
    cycling = pd.concat(lis, axis=0 , ignore_index=True)
except Exception as e:
    lg.error(e)

b'Skipping line 485: expected 7 fields, saw 8\n'
b'Skipping line 485: expected 7 fields, saw 8\n'

cycling.shape

(7198, 7)

cycling.dtypes

0      int64
1    float64
2    float64
3    float64
4    float64
5    float64
6    float64
dtype: object

cycling.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 7198 entries, 0 to 7197
Data columns (total 7 columns):
 #   Column  Non-Null Count  Dtype  
---  ------  --------------  -----  
 0   0       7198 non-null   int64  
 1   1       7198 non-null   float64
 2   2       7198 non-null   float64
 3   3       7198 non-null   float64
 4   4       7198 non-null   float64
 5   5       7198 non-null   float64
 6   6       7198 non-null   float64
dtypes: float64(6), int64(1)
memory usage: 393.8 KB

cycling.describe()

	0	1	2	3	4	5	6
count	7198.000000	7198.000000	7198.000000	7198.000000	7198.000000	7198.000000	7198.000000
mean	59858.363434	36.747305	2.688654	16.395645	2.989557	17.397501	3.037647
std	34633.776409	3.821483	1.934684	3.605867	1.708674	3.302723	1.765061
min	0.000000	18.500000	0.000000	0.000000	0.000000	0.000000	0.000000
25%	29812.500000	34.250000	1.300000	14.000000	1.660000	15.250000	1.700000
50%	59750.000000	36.250000	2.170000	16.330000	2.770000	17.500000	2.740000
75%	89750.000000	40.000000	3.700000	18.750000	4.030000	19.750000	4.030000
max	119750.000000	45.000000	14.170000	28.500000	10.610000	27.000000	11.150000

cycling.isnull().sum()

0    0
1    0
2    0
3    0
4    0
5    0
6    0
dtype: int64

try:
    lg.info("adding the label")
    cycling['label'] = 'cycling'
except Exception as e:
    lg.error(e)

cycling.head()

	0	1	2	3	4	5	6	label
0	0	32.0	4.85	17.50	3.35	22.50	3.20	cycling
1	250	40.5	1.12	14.00	2.24	21.75	1.30	cycling
2	500	40.5	2.60	11.33	4.50	18.25	5.31	cycling
3	750	34.5	1.50	20.67	2.87	19.00	2.83	cycling
4	1000	34.5	1.50	21.25	3.27	18.25	4.38	cycling

try:
    lg.info("changing the directory lying folder")
    import glob
    path = r'C:\Users\sid\Downloads\AReM\lying'
    all_csv = glob.glob(path + "/*.csv")
    
    lis = []
    for csv in all_csv:
        df = pd.read_csv(csv , sep=',' , header=None, skiprows=5,  error_bad_lines=False)
        lis.append(df)
    
    lying = pd.concat(lis, axis=0 , ignore_index=True)
except Exception as e:
    lg.error(e)

try:
    lg.info("adding the label for activity lying")
    lying['label'] = 'lying'
except Exception as e:
    lg.error(e)

lying.shape

(7200, 8)

lying.head()

	0	1	2	3	4	5	6	label
0	0	29.0	0.0	9.00	0.71	8.50	0.50	lying
1	250	29.0	0.0	8.00	0.71	8.50	0.50	lying
2	500	29.0	0.0	8.00	0.71	8.00	1.00	lying
3	750	28.5	0.5	8.25	0.43	8.75	0.43	lying
4	1000	29.0	0.0	8.75	1.09	9.00	0.00	lying

lying.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 7200 entries, 0 to 7199
Data columns (total 8 columns):
 #   Column  Non-Null Count  Dtype  
---  ------  --------------  -----  
 0   0       7200 non-null   int64  
 1   1       7200 non-null   float64
 2   2       7200 non-null   float64
 3   3       7200 non-null   float64
 4   4       7200 non-null   float64
 5   5       7200 non-null   float64
 6   6       7200 non-null   float64
 7   label   7200 non-null   object 
dtypes: float64(6), int64(1), object(1)
memory usage: 450.1+ KB

lying.isnull().sum()

0        0
1        0
2        0
3        0
4        0
5        0
6        0
label    0
dtype: int64

try:
    lg.info("combining all files in the sitting folder")
    path = r'C:\Users\sid\Downloads\AReM\sitting'
    all_csv = glob.glob(path + "/*.csv")
    
    lis = []
    
    for i in all_csv:
        df = pd.read_csv(csv , sep=',' , header=None, skiprows=5,  error_bad_lines=False)
        lis.append(df)
    sitting = pd.concat(lis, axis=0, ignore_index=True)
except Exception as e:
    lg.error(e)

try:
    lg.info("adding the label for activity sitting")
    sitting['label'] = 'sitting'
except Exception as e:
    lg.error(e)

sitting.isnull().sum()

0        0
1        0
2        0
3        0
4        0
5        0
6        0
label    0
dtype: int64

sitting.dtypes

0          int64
1        float64
2        float64
3        float64
4        float64
5        float64
6        float64
label     object
dtype: object

try:
    lg.info("changing the folder to standing and combining the datasets")
    path = r'C:/Users/sid/Downloads/AReM/standing'
    all_csv = glob.glob(path + "/*.csv")
    
    lis = []
    
    for i in all_csv:
        df = pd.read_csv(csv , sep=',' , header=None, skiprows=5,  error_bad_lines=False)
        lis.append(df)
    
    standing = pd.concat(lis, axis=0, ignore_index=True)

except Exception as e:
    lg.error(e)

standing.isnull().sum()

0    0
1    0
2    0
3    0
4    0
5    0
6    0
dtype: int64

try:
    lg.info("adding the label for activity sitting")
    standing['label'] = 'standing'
except Exception as e:
    lg.error(e)

standing.isnull().sum()

0        0
1        0
2        0
3        0
4        0
5        0
6        0
label    0
dtype: int64

try:
    lg.info("changing the folder and combining the datasets")
    path = r'C:/Users/sid/Downloads/AReM/walking'
    all_csv = glob.glob(path + "/*.csv")
    
    lis = []
    
    for i in all_csv:
        df = pd.read_csv(csv , sep=',' , header=None, skiprows=5,  error_bad_lines=False)
        lis.append(df)
    
    walking = pd.concat(lis, axis=0, ignore_index=True)

except Exception as e:
    lg.error(e)    

walking.isnull().sum()

0    0
1    0
2    0
3    0
4    0
5    0
6    0
dtype: int64

try:
    lg.info("adding the label for walking")
    walking['label'] = 'walking'
except Exception as e:
    lg.error(e)

walking

	0	1	2	3	4	5	6	label
0	0	39.67	0.47	6.00	0.82	16.00	1.00	walking
1	250	39.50	0.50	4.50	1.12	17.75	0.43	walking
2	500	39.67	0.47	4.50	1.12	18.33	0.47	walking
3	750	39.50	0.50	4.50	2.06	17.50	0.50	walking
4	1000	39.33	0.47	5.33	0.47	17.00	0.71	walking
...	...	...	...	...	...	...	...	...
7195	118750	41.50	0.50	10.67	0.47	14.00	0.82	walking
7196	119000	41.50	0.50	10.80	0.40	14.40	0.80	walking
7197	119250	41.75	0.43	10.00	0.00	13.67	0.94	walking
7198	119500	42.00	0.00	9.40	0.49	14.00	1.10	walking
7199	119750	41.75	0.43	9.00	0.00	14.33	0.47	walking

7200 rows × 8 columns

try:
    lg.info("combining all the dataset into one")
    final_df = pd.concat([bending1, df_bending2, cycling, lying, sitting , standing, walking])
except Exception as e:
    lg.error(e)

try:
    lg.info("adding the columns")
    final_df.columns = ['time', 'avg_rss12', 'var_rss12', 'avg_rss13', 'var_rss13', 'avg_rss23',
       'var_rss23', 'label']
except Exception as e:
    lg.error(e)

final_df.isnull().sum()

time         0
avg_rss12    0
var_rss12    0
avg_rss13    0
var_rss13    0
avg_rss23    0
var_rss23    0
label        0
dtype: int64

final_df.dtypes

time           int64
avg_rss12    float64
var_rss12    float64
avg_rss13    float64
var_rss13    float64
avg_rss23    float64
var_rss23    float64
label         object
dtype: object

ProfileReport(final_df)

%matplotlib inline


import warnings
warnings.filterwarnings('ignore')

plt.figure(figsize=(12, 10))
plt.style.use('fivethirtyeight')
sns.distplot(final_df['time'])

<AxesSubplot:xlabel='time', ylabel='Density'>

import warnings
warnings.filterwarnings('ignore')

plt.figure(figsize=(12, 10))
plt.style.use('fivethirtyeight')
sns.distplot(final_df['avg_rss12'])

<AxesSubplot:xlabel='avg_rss12', ylabel='Density'>

import warnings
warnings.filterwarnings('ignore')

plt.figure(figsize=(12, 10))
plt.style.use('fivethirtyeight')
sns.distplot(final_df['var_rss12'])

<AxesSubplot:xlabel='var_rss12', ylabel='Density'>

import warnings
warnings.filterwarnings('ignore')

plt.figure(figsize=(12, 10))
plt.style.use('fivethirtyeight')
sns.distplot(final_df['avg_rss13'])

<AxesSubplot:xlabel='avg_rss13', ylabel='Density'>

import warnings
warnings.filterwarnings('ignore')

plt.figure(figsize=(12, 10))
plt.style.use('fivethirtyeight')
sns.distplot(final_df['var_rss13'])

<AxesSubplot:xlabel='var_rss13', ylabel='Density'>

import warnings
warnings.filterwarnings('ignore')

plt.figure(figsize=(12, 10))
plt.style.use('fivethirtyeight')
sns.distplot(final_df['avg_rss23'])

<AxesSubplot:xlabel='avg_rss23', ylabel='Density'>

import warnings
warnings.filterwarnings('ignore')

plt.figure(figsize=(12, 10))
plt.style.use('fivethirtyeight')
sns.distplot(final_df['var_rss23'])

<AxesSubplot:xlabel='var_rss23', ylabel='Density'>

final_df.columns

Index(['time', 'avg_rss12', 'var_rss12', 'avg_rss13', 'var_rss13', 'avg_rss23',
       'var_rss23', 'label'],
      dtype='object')

plt.figure(figsize=(12, 10))
sns.countplot(final_df['label'])

<AxesSubplot:xlabel='label', ylabel='count'>

final_df

	time	avg_rss12	var_rss12	avg_rss13	var_rss13	avg_rss23	var_rss23	label
0	0	39.25	0.43	22.75	0.43	33.75	1.30	bending1
1	250	39.25	0.43	23.00	0.00	33.00	0.00	bending1
2	500	39.25	0.43	23.25	0.43	33.00	0.00	bending1
3	750	39.50	0.50	23.00	0.71	33.00	0.00	bending1
4	1000	39.50	0.50	24.00	0.00	33.00	0.00	bending1
...	...	...	...	...	...	...	...	...
7195	118750	41.50	0.50	10.67	0.47	14.00	0.82	walking
7196	119000	41.50	0.50	10.80	0.40	14.40	0.80	walking
7197	119250	41.75	0.43	10.00	0.00	13.67	0.94	walking
7198	119500	42.00	0.00	9.40	0.49	14.00	1.10	walking
7199	119750	41.75	0.43	9.00	0.00	14.33	0.47	walking

42238 rows × 8 columns

y = final_df['label']

from sklearn import preprocessing
le = preprocessing.LabelEncoder()

target = le.fit_transform(y)

target.shape

(42238,)

x = final_df.drop('label', axis=1)

try:
    lg.info("standardize the data")
    scalar = StandardScaler()
    ProfileReport(pd.DataFrame(scalar.fit_transform(x)))
    x_scaled = scalar.fit_transform(x)
except Exception as e:
    lg.error(e)

x_scaled

array([[-1.72844562, -0.4276623 , -0.22750026, ..., -0.52477425,
         2.68657524,  0.12827614],
       [-1.72122838, -0.4276623 , -0.22750026, ..., -0.84949444,
         2.57595644, -0.87013852],
       [-1.71401115, -0.4276623 , -0.22750026, ..., -0.52477425,
         2.57595644, -0.87013852],
       ...,
       [ 1.71417484, -0.11381752, -0.22750026, ..., -0.84949444,
        -0.2750589 , -0.14820792],
       [ 1.72139208, -0.08243304, -0.5606259 , ..., -0.47946446,
        -0.22638662, -0.02532612],
       [ 1.72860931, -0.11381752, -0.22750026, ..., -0.84949444,
        -0.17771435, -0.50917322]])

def vif_score(x):
    try: 
        lg.info("vif score ")
        scaler = StandardScaler()
        arr = scaler.fit_transform(x)
        return pd.DataFrame([[x.columns[i], variance_inflation_factor(arr,i)] for i in range(arr.shape[1])], columns=["FEATURE", "VIF_SCORE"])
    except Exception as e:
        lg.error(e)

vif_score(x)

	FEATURE	VIF_SCORE
0	time	1.110289
1	avg_rss12	1.248030
2	var_rss12	1.768504
3	avg_rss13	1.995340
4	var_rss13	1.578514
5	avg_rss23	1.752785
6	var_rss23	1.651067

splitting the dataset

try:
    lg.info("splitting the training and testing")
    x_train, x_test, y_train, y_test = train_test_split(x_scaled , target , test_size = .20 , random_state = 144)
except Exception as e:
    lg.error(e)

logr_liblinear = LogisticRegression(verbose=1,solver='liblinear')

logr_liblinear.fit(x_train,y_train )

[LibLinear]

LogisticRegression(solver='liblinear', verbose=1)

y_pred = logr_liblinear.predict(x_test)

#print the scores on training and test set

print('Training set score: {:.4f}'.format(logr_liblinear.score(x_train, y_train)))

print('Test set score: {:.4f}'.format(logr_liblinear.score(x_test, y_test)))

Training set score: 0.5363
Test set score: 0.5320

# Print the Confusion Matrix and slice it into four pieces

from sklearn.metrics import confusion_matrix

cm = confusion_matrix(y_test, y_pred)

print('Confusion matrix\n\n', cm)

print('\nTrue Positives(TP) = ', cm[0,0])

print('\nTrue Negatives(TN) = ', cm[1,1])

print('\nFalse Positives(FP) = ', cm[0,1])

print('\nFalse Negatives(FN) = ', cm[1,0])

Confusion matrix

 [[ 582    1   21    2   16    3   16]
 [  40  426   58   42    5    1    2]
 [   8   25 1354    6   19    3    4]
 [   0   36   11 1239  172   19   59]
 [   0    0   60  438  366  130  392]
 [   0    0   42  463  413  124  418]
 [   0    0   60  457  398  114  403]]

True Positives(TP) =  582

True Negatives(TN) =  426

False Positives(FP) =  1

False Negatives(FN) =  40

# visualize confusion matrix with seaborn heatmap
from sklearn.metrics import classification_report
print(classification_report(y_test, y_pred))

cm_matrix = pd.DataFrame(data=cm)
plt.rcParams['figure.figsize'] = (5, 5)

sns.heatmap(cm_matrix, annot=True, fmt='d', cmap='YlGnBu')

              precision    recall  f1-score   support

           0       0.92      0.91      0.92       641
           1       0.87      0.74      0.80       574
           2       0.84      0.95      0.90      1419
           3       0.47      0.81      0.59      1536
           4       0.26      0.26      0.26      1386
           5       0.31      0.08      0.13      1460
           6       0.31      0.28      0.30      1432

    accuracy                           0.53      8448
   macro avg       0.57      0.58      0.56      8448
weighted avg       0.51      0.53      0.50      8448

<AxesSubplot:>

log_newton = LogisticRegression(verbose=1, solver='newton-cg')

log_newton.fit(x_train, y_train)

[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    2.0s finished

LogisticRegression(solver='newton-cg', verbose=1)

y_pred = log_newton.predict(x_test)

print('Training-set accuracy score: {0:0.4f}'. format(accuracy_score(y_test, y_pred)))

Training-set accuracy score: 0.5397

#print the scores on training and test set

print('Training set score: {:.4f}'.format(log_newton.score(x_train, y_train)))

print('Test set score: {:.4f}'.format(log_newton.score(x_test, y_test)))

Training set score: 0.5447
Test set score: 0.5397

# Print the Confusion Matrix and slice it into four pieces

from sklearn.metrics import confusion_matrix

cm = confusion_matrix(y_test, y_pred)

print('Confusion matrix\n\n', cm)

print('\nTrue Positives(TP) = ', cm[0,0])

print('\nTrue Negatives(TN) = ', cm[1,1])

print('\nFalse Positives(FP) = ', cm[0,1])

print('\nFalse Negatives(FN) = ', cm[1,0])

Confusion matrix

 [[ 598    6   33    0    2    0    2]
 [  40  455   49   28    0    2    0]
 [  10   36 1354    4    3   11    1]
 [   0   45    5 1205  188   34   59]
 [   0    0   13  414  387  183  389]
 [   0    0   14  432  436  158  420]
 [   0    0   20  433  415  162  402]]

True Positives(TP) =  598

True Negatives(TN) =  455

False Positives(FP) =  6

False Negatives(FN) =  40

from sklearn.metrics import classification_report

print(classification_report(y_test, y_pred))

cm_matrix = pd.DataFrame(data=cm)
plt.rcParams['figure.figsize'] = (5, 5)

sns.heatmap(cm_matrix, annot=True, fmt='d', cmap='YlGnBu')

              precision    recall  f1-score   support

           0       0.92      0.93      0.93       641
           1       0.84      0.79      0.82       574
           2       0.91      0.95      0.93      1419
           3       0.48      0.78      0.59      1536
           4       0.27      0.28      0.27      1386
           5       0.29      0.11      0.16      1460
           6       0.32      0.28      0.30      1432

    accuracy                           0.54      8448
   macro avg       0.57      0.59      0.57      8448
weighted avg       0.51      0.54      0.51      8448

<AxesSubplot:>

logr_sag = LogisticRegression(verbose=1, solver='sag')

logr_sag.fit(x_train, y_train)

[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.

convergence after 19 epochs took 1 seconds

[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.4s finished

LogisticRegression(solver='sag', verbose=1)

y_pred = logr_sag.predict(x_test)
y_pred

array([2, 4, 6, ..., 3, 4, 2])

#print the scores on training and test set

print('Training set score: {:.4f}'.format(logr_sag.score(x_train, y_train)))

print('Test set score: {:.4f}'.format(logr_sag.score(x_test, y_test)))

Training set score: 0.5447
Test set score: 0.5397

# Print the Confusion Matrix and slice it into four pieces

from sklearn.metrics import confusion_matrix

cm = confusion_matrix(y_test, y_pred)

print('Confusion matrix\n\n', cm)

print('\nTrue Positives(TP) = ', cm[0,0])

print('\nTrue Negatives(TN) = ', cm[1,1])

print('\nFalse Positives(FP) = ', cm[0,1])

print('\nFalse Negatives(FN) = ', cm[1,0])

Confusion matrix

 [[ 598    6   33    0    2    0    2]
 [  40  455   49   28    0    2    0]
 [  10   36 1354    4    3   11    1]
 [   0   45    5 1205  188   34   59]
 [   0    0   13  414  387  183  389]
 [   0    0   14  432  436  158  420]
 [   0    0   20  433  415  162  402]]

True Positives(TP) =  598

True Negatives(TN) =  455

False Positives(FP) =  6

False Negatives(FN) =  40

from sklearn.metrics import classification_report

print(classification_report(y_test, y_pred))

cm_matrix = pd.DataFrame(data=cm)
plt.rcParams['figure.figsize'] = (5, 5)

sns.heatmap(cm_matrix, annot=True, fmt='d', cmap='YlGnBu')

              precision    recall  f1-score   support

           0       0.92      0.93      0.93       641
           1       0.84      0.79      0.82       574
           2       0.91      0.95      0.93      1419
           3       0.48      0.78      0.59      1536
           4       0.27      0.28      0.27      1386
           5       0.29      0.11      0.16      1460
           6       0.32      0.28      0.30      1432

    accuracy                           0.54      8448
   macro avg       0.57      0.59      0.57      8448
weighted avg       0.51      0.54      0.51      8448

<AxesSubplot:>

logr_saga = LogisticRegression(verbose=1, solver='saga')

logr_saga.fit(x_train , y_train)

[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.

convergence after 13 epochs took 1 seconds

[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.3s finished

LogisticRegression(solver='saga', verbose=1)

y_pred = logr_saga.predict(x_test)

#print the scores on training and test set

print('Training set score: {:.4f}'.format(logr_saga.score(x_train, y_train)))

print('Test set score: {:.4f}'.format(logr_saga.score(x_test, y_test)))

Training set score: 0.5449
Test set score: 0.5389

# Print the Confusion Matrix and slice it into four pieces

from sklearn.metrics import confusion_matrix

cm = confusion_matrix(y_test, y_pred)

print('Confusion matrix\n\n', cm)

print('\nTrue Positives(TP) = ', cm[0,0])

print('\nTrue Negatives(TN) = ', cm[1,1])

print('\nFalse Positives(FP) = ', cm[0,1])

print('\nFalse Negatives(FN) = ', cm[1,0])

Confusion matrix

 [[ 598    6   33    0    2    0    2]
 [  40  455   49   28    0    2    0]
 [  10   36 1354    4    3   11    1]
 [   0   45    5 1205  191   34   56]
 [   0    0   13  414  390  183  386]
 [   0    0   14  432  442  158  414]
 [   0    0   20  433  424  162  393]]

True Positives(TP) =  598

True Negatives(TN) =  455

False Positives(FP) =  6

False Negatives(FN) =  40

from sklearn.metrics import classification_report

print(classification_report(y_test, y_pred))

cm_matrix = pd.DataFrame(data=cm)
plt.rcParams['figure.figsize'] = (5, 5)

sns.heatmap(cm_matrix, annot=True, fmt='d', cmap='YlGnBu')

              precision    recall  f1-score   support

           0       0.92      0.93      0.93       641
           1       0.84      0.79      0.82       574
           2       0.91      0.95      0.93      1419
           3       0.48      0.78      0.59      1536
           4       0.27      0.28      0.27      1386
           5       0.29      0.11      0.16      1460
           6       0.31      0.27      0.29      1432

    accuracy                           0.54      8448
   macro avg       0.57      0.59      0.57      8448
weighted avg       0.51      0.54      0.51      8448

<AxesSubplot:>

logr_lbfgs = LogisticRegression(verbose=1, solver='lbfgs')

logr_lbfgs.fit(x_train, y_train)

[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    1.3s finished

LogisticRegression(verbose=1)

y_pred = logr_lbfgs.predict(x_test)

from sklearn.metrics import accuracy_score
print("model accuracy score: {0:0.4f}".format(accuracy_score(y_test, y_pred)))

model accuracy score: 0.5397

#print the scores on training and test set

print('Training set score: {:.4f}'.format(logr_lbfgs.score(x_train, y_train)))

print('Test set score: {:.4f}'.format(logr_lbfgs.score(x_test, y_test)))

Training set score: 0.5447
Test set score: 0.5397

# Print the Confusion Matrix and slice it into four pieces

from sklearn.metrics import confusion_matrix

cm = confusion_matrix(y_test, y_pred)

print('Confusion matrix\n\n', cm)

print('\nTrue Positives(TP) = ', cm[0,0])

print('\nTrue Negatives(TN) = ', cm[1,1])

print('\nFalse Positives(FP) = ', cm[0,1])

print('\nFalse Negatives(FN) = ', cm[1,0])

Confusion matrix

 [[ 598    6   33    0    2    0    2]
 [  40  455   49   28    0    2    0]
 [  10   36 1354    4    3   11    1]
 [   0   45    5 1205  188   34   59]
 [   0    0   13  414  387  183  389]
 [   0    0   14  432  436  158  420]
 [   0    0   20  433  415  162  402]]

True Positives(TP) =  598

True Negatives(TN) =  455

False Positives(FP) =  6

False Negatives(FN) =  40

# visualize confusion matrix with seaborn heatmap

from sklearn.metrics import classification_report

print(classification_report(y_test, y_pred))

cm_matrix = pd.DataFrame(data=cm)
plt.rcParams['figure.figsize'] = (5, 5)

sns.heatmap(cm_matrix, annot=True, fmt='d', cmap='YlGnBu')

              precision    recall  f1-score   support

           0       0.92      0.93      0.93       641
           1       0.84      0.79      0.82       574
           2       0.91      0.95      0.93      1419
           3       0.48      0.78      0.59      1536
           4       0.27      0.28      0.27      1386
           5       0.29      0.11      0.16      1460
           6       0.32      0.28      0.30      1432

    accuracy                           0.54      8448
   macro avg       0.57      0.59      0.57      8448
weighted avg       0.51      0.54      0.51      8448

              precision    recall  f1-score   support

           0       0.92      0.93      0.93       641
           1       0.84      0.79      0.82       574
           2       0.91      0.95      0.93      1419
           3       0.48      0.78      0.59      1536
           4       0.27      0.28      0.27      1386
           5       0.29      0.11      0.16      1460
           6       0.32      0.28      0.30      1432

    accuracy                           0.54      8448
   macro avg       0.57      0.59      0.57      8448
weighted avg       0.51      0.54      0.51      8448

<AxesSubplot:>